During the past decade, the incidence of adenocarcinomas of the esophagus and gastric, cardia has increased markedly in the U.S., particularly among white men. In contrast, the rates of squamous cell carcinoma of the esophagus and other more distal stomach cancers have remained fairly stable. This proposal addresses possible reasons for the rising incidence of these tumors. Specifically, we propose to conduct a population-based case-control study of esophageal and gastric carcinomas (N=316) in men and women of all races who are between 30-74 years of age at the time of diagnosis, using the western Washington cancer registry. Controls (N=316) of similar age and gender from the same geographic area will be selected by random digit dialing. We will examine relative risks associated with: i) medications that inhibit gastric acid secretion, neutralize gastric acid, or irritate gastric/esophageal mucosa, ii) Barrett's esophagus and other benign gastrointestinal conditions, and iii) serologic evidence of prior infection with Helicobacter pylori. We also will collect information on other identified and hypothesized risk factors for cancers of the esophagus and stomach, including smoking, alcohol use, dietary intake, occupational history, and family history of gastrointestinal cancers. These and other exposures will be analyzed as independent predictors of risk in order to determine whether there are different risk factor profiles for adenocarcinomas of the esophagus/gastric cardia compared to other carcinomas of the esophagus and stomach, and as confounding variables in relation to the above specific aims. We will conduct in-person interviews, and obtain additional information from medical and pharmacy records. Cases and controls will be asked to provide a blood sample that will be used to measure antibodies to Helicobacter pylori, and for storage of leukocytes for future cytogenetic studies. Tumor tissue slides and blocks also will be obtained for all cases for uniform confirmation of histological diagnosis, determination of H. pylori infection status in tumor tissue, and future determination of specific oncogenes that may be associated with development of these adenocarcinomas.